File size: 3,391 Bytes
3ee94dd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c68294e
 
 
3ee94dd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
import tempfile
from typing import Optional
from TTS.config import load_config
import gradio as gr
import numpy as np
from TTS.utils.manage import ModelManager
from TTS.utils.synthesizer import Synthesizer


MODELS = {}
SPEAKERS = {}


manager = ModelManager()
MODEL_NAMES = manager.list_tts_models()

# filter out multi-speaker models
filters = ["vctk", "your_tts"]
MODEL_NAMES = [model_name for model_name in MODEL_NAMES if not any(f in model_name for f in filters)]

# reorder models
MODEL_NAMES[0], MODEL_NAMES[1], MODEL_NAMES[2]= MODEL_NAMES[1], MODEL_NAMES[2], MODEL_NAMES[0]
print(MODEL_NAMES)


def tts(text: str, model_name: str, speaker_idx: str=None):
    print(text, model_name)
    # download model
    model_path, config_path, model_item = manager.download_model(f"tts_models/{model_name}")
    vocoder_name: Optional[str] = model_item["default_vocoder"]
    # download vocoder
    vocoder_path = None
    vocoder_config_path = None
    if vocoder_name is not None:
        vocoder_path, vocoder_config_path, _ = manager.download_model(vocoder_name)
    # init synthesizer
    synthesizer = Synthesizer(
        model_path, config_path, None, None, vocoder_path, vocoder_config_path,
    )
    # synthesize
    if synthesizer is None:
        raise NameError("model not found")
    wavs = synthesizer.tts(text, speaker_idx)
    # return output
    with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as fp:
        synthesizer.save_wav(wavs, fp)
        return fp.name



article= """
### Visit us on <a href=https://coqui.ai>Coqui.ai</a> and drop a 🌟 to <a href=https://github.com/coqui-ai/TTS>CoquiTTS</a>.

### You can run CoquiTTS on your machine. Check out our <a href="https://tts.readthedocs.io/en/latest/inference.html">documentation</a>.


```bash
$ pip install TTS
...
$ tts --list_models
...
$ tts --text "Text for TTS" --model_name "<type>/<language>/<dataset>/<model_name>" --out_path folder/to/save/output.wav
```

### 👑 Model contributors

- <a href="https://github.com/nmstoker/">@nmstoker</a>
- <a href="https://github.com/kaiidams/">@kaiidams</a>
- <a href="https://github.com/WeberJulian/">@WeberJulian,</a>
- <a href="https://github.com/Edresson/">@Edresson</a>
- <a href="https://github.com/thorstenMueller/">@thorstenMueller</a>
- <a href="https://github.com/r-dh/">@r-dh</a>
- <a href="https://github.com/kirianguiller/">@kirianguiller</a>
- <a href="https://github.com/robinhad/">@robinhad</a>

Drop a ✨PR✨ on 🐸TTS to share a new model and have it included here.
"""

iface = gr.Interface(
    fn=tts,
    inputs=[
        gr.inputs.Textbox(
            label="Input",
            default="Hello, how are you?",
        ),
        gr.inputs.Radio(
            label="Pick a TTS Model",
            choices=MODEL_NAMES,
        ),
        # gr.inputs.Dropdown(label="Select a speaker", choices=SPEAKERS, default=None)
        # gr.inputs.Audio(source="microphone", label="Record your voice.", type="numpy", label=None, optional=False)
    ],
    outputs=gr.outputs.Audio(label="Output"),
    title="🐸💬 CoquiTTS Demo",
    theme="grass",
    description="🐸💬  Coqui TTS - a deep learning toolkit for Text-to-Speech, battle-tested in research and production.",
    article=article,
    allow_flagging=False,
    flagging_options=['error', 'bad-quality', 'wrong-pronounciation'],
    live=False
)
iface.launch(share=False)